Thursday, November 12, 2015: 10:30 AM
355B (Salt Palace Convention Center)
Density functional theory (DFT) has matured into a powerful practical tool that may be straightforwardly treated as a black box in order to elucidate and interpret trends in chemical and physical bonding interactions in organic systems. However, when mid-row, open shell transition metals such as iron participate in bonding either in inorganic chemistry or in correlated materials, magnetic ordering or spin state preference are known to be highly sensitive to the choice of exchange-correlation functional. Here, we aim to estimate the range of predictions that can be obtained for representative transition metal materials when taking into account the most likely decisions practitioners make in setting up a DFT calculation. We exploit the adiabatic connection to interpret the relationship between the typically thought of as delocalization-driven low-spin preference of pure generalized gradient approximation functionals to the high-spin preference in common hybrid functionals. Using these trends, we identify robust linear scaling relationships that correlate to direct metal-ligand interactions and elucidate how sensitive any given pair of iron oxidation states and ligands is to the extent of exact exchange in a functional. Importantly, we identify a counterintuitive relationship between the degree of charge localized on a metal center for a given spin state and that state’s relative stability when treatments of exchange are varied.